fixed bugs on descentParafoil, there were some mismatches between the simulink...

fixed bugs on descentParafoil, there were some mismatches between the simulink model and the ode, but still something goes wrong, very likely with the quaternions. In addition, now the GPS works also in descent, but this brings to light other problems in the attitude estimation, again, the reason has to be found in the quaternion prediction/correction scheme